MIMaL

MIMaL infers protein–metabolite regulatory relationships by integrating proteomics and metabolomics data and applying machine-learning model interpretation to quantify protein control (ProC) over metabolites.

Key Features:

• Integration of Multi-Omic Data: Combines proteomics and metabolomics datasets to provide a holistic view of biological states and interactions within cellular environments.
• Machine Learning and Model Interpretation: Applies machine learning algorithms with model interpretation strategies to identify connections across omic layers and quantify protein control (ProC) over metabolites.
• Discovery of Novel Protein Regulators: Maps proteins with inferred ProC onto known genetic and metabolic networks to identify novel protein regulators of metabolites.
• Functional Prediction and Validation: Clusters ProC magnitudes across metabolites to predict gene functions and has predicted and validated roles for genes such as YJR120W and YDL157C while providing insights into SDH9, ISC1, and FMP52.

Scientific Applications:

• Multi-omic Integration: Enables coordinated analysis of proteomics and metabolomics data to study molecular layer interactions.
• Regulatory Mechanism Discovery: Reveals regulatory mechanisms by linking protein control to metabolic changes within genetic and metabolic network contexts.
• Gene Function Prediction and Validation: Predicts gene functions via clustering of ProC magnitudes and supports experimental validation of uncharacterized genes.
• Hypothesis Generation and Testing: Generates hypotheses about protein–metabolite relationships for downstream experimental testing.

Methodology:

Integrates proteomics and metabolomics datasets; applies machine learning algorithms with model interpretation to compute protein control (ProC) over metabolites; maps inferred proteins onto genetic and metabolic networks; clusters ProC magnitudes across metabolites for functional prediction.